Riparian Zone Variability In the Foothills of Alberta
| Abstract | Introduction | Data | Methods | Results | Discussion | Sources of error | References |
|
Defining the riparian zone Because woody debris input is controlled primarily by riparian forest, I limited my analysis to these zones. But how deep is the riparian zone exactly? A typical and commonly used measure of riparian zones is one tree length from the stream edge.4,5 However, some suggest the width should be 1.1 maximum tree heights are more realistic.5 In Alberta, riparian forests are legally defined as 30m deep for most streams. I used this definition of the riparian zone because it is familiar and comparable to other studies, and because it is deeper than 1.1 maximum tree heights in these forests where coniferous trees rarely exceed 20m. Creating riparian zones Stream reach lines were adjusted up to 100m. This produced a line smooth enough to generate rectangular riparian zone polygons. The original stream attribute table remained intact in this process. The smoothed reach lines were buffered using a 30m buffer on both sides of the line, with a flat edge. Separate buffer polygons were maintained for each stream segment by choosing the “no dissolve” option, producing around 13 000 polygons per watershed. The same procedure was performed to create 2m buffers of the streams, as a proxy for streambed elevation.  Topographic roughness Zonal statistics were calculated from the DEM for each 30m and 2m riparian polygon. The standard deviation was appended to the polygon attribute table, and used as a measure of variability. Forest analysis The Vegetation Index map was edited to remove all non-forested areas. To do so, an SQL statement was used removing polygons with stand origin dates of zero. Next, all forest polygons established after 1910 were removed. Stands with dominant tree species other than spruce were removed. Finally, all areas overlapping with post-1907 forest fires and cutblock polygons were erased, producing a layer with only old growth, undisturbed spruce forest. This layer was converted to raster format with a cell size of 4m. Zonal statistics were calculated from the old growth raster file within each riparian polygon. If most cells in the riparian zone were spruce, the polygon was considered similar to my study sites. Any parts of the stream network that were outside the boundaries of the Vegetation Index were clipped.   Determining relationships Correlations were performed between variables using the Correlate function in Excel. Back to top Results |